converting csv timestamp to datetime on import

I have a CSV that has two columns, Date and Time. The general format of the text is colum 1, Date, is m/DD/YYYY, and column2, Time, is HH:mm:ss SSS. However, the output device outputs the text for time omitting the zero in cases where the digit is not bordered by other numbers in its value category. For example, one case reads out:

HH:m:ss S (11:1:30 1)

when it should read HH:mm:ss SSS (11:01:30 001) if it generated equal length strings.

I cannot change the output device to correct this to always include 'HH:mm:ss SSS' to perform a simple string input to use in the datetime(DateStrings,'InputFormat',infmt) method as the string changes length.

Answer 1

[Stephen23]

type test.csv

6/28/2022,11:1:30 1
06/28/2022,11:01:30 001

T = readtable('test.csv','Format','%{M/d/y}D%T%d','Delimiter',{',',' '})

T = 2×3 table
      Var1         Var2      Var3
    _________    ________    ____

    6/28/2022    11:01:30     1  
    6/28/2022    11:01:30     1  

D = T.Var1 + T.Var2 + milliseconds(T.Var3);
D.Format = 'yyyy-MM-dd HH:mm:ss.SSS'

D = 2×1 datetime array
   2022-06-28 11:01:30.001
   2022-06-28 11:01:30.001

[Russell Mott]

Thank you for that answer. in the read table function, I see some guidance on how the 'Format' name value pair works, but it doesnt seem like a complete guide. do you have any links to resources to understand the full functionality of the 'Format' option?

[Stephen23]

The READTABLE documentation states for the FORMAT that "The conversion specifiers are the same as the specifiers accepted by the textscan function", so this would be the resource to read:

https://www.mathworks.com/help/matlab/ref/textscan.html#btghhyz-1-formatSpec